Lung mast cell density defines a subpopulation of patients with idiopathic pulmonary fibrosis

Cha S‐I, Chang C S, Kim E K, Lee J W, Matthay M A, Golden J A, Elicker B M, Jones K, Collard H R & Wolters P J (2012) Histopathology  61, 98–106 Lung mast cell density defines a subpopulation of patients with idiopathic pulmonary fibrosis Aims: The relationship of mast cells to the pathogenesis of lung fibrosis remains undefined despite recognition of their presence in the lungs of patients with pulmonary fibrosis. This study was performed to characterize the relationship of mast cells to fibrotic lung diseases. Methods and results: Lung tissues from patients with idiopathic pulmonary fibrosis (IPF), chronic hypersensitivity pneumonitis (HP), systemic sclerosis (SSc)‐related interstitial lung disease (ILD) and normal individuals were subjected to chymase immunostaining and the mast cell density quantified. Eosinophils were quantified by immunostaining for eosinophil peroxidase. Changes in lung function were correlated with mast cell density. Lung tissue obtained from IPF patients had a higher density of chymase‐immunoreactive mast cells than that from patients with HP, SSc‐related ILD or normal lungs. IPF lung tissue had a higher density of eosinophils than normal lung. There was no correlation between mast cell density and eosinophil density in IPF lung. IPF patients with high mast cell density had a slower rate of decline in forced vital capacity (FVC) than IPF patients with low mast cell density. Conclusions: Mast cell density in IPF lungs is higher than in other fibrotic lung diseases and normal lungs. Increased mast cell density in IPF may predict slower disease progression.     

Comparative decellularization and recellularization of normal versus emphysematous human lungs

Acellular whole human lung scaffolds represent a unique opportunity for ex vivo tissue engineering. However, it remains unclear whether lungs from individuals with chronic lung diseases such as chronic obstructive pulmonary disease (COPD) can be appropriately decellularized and recellularized. To assess this, cadaveric human lungs from normal (non-smoking) patients and from patients with COPD (smoking history) were decellularized and found by histochemical and immunohistochemical staining, electron microscopy, and mass spectrometry to retain characteristic histological architecture and extracellular matrix components (ECM) reflecting either normal or COPD, particularly emphysematous, origin. Inoculation of human bronchial epithelial cells, endothelial progenitor cells, bone marrow-derived mesenchymal stem cells, and lung fibroblasts via airway or vascular routes into small, excised segments of the decellularized lungs demonstrated that normal lung scaffolds robustly supported initial engraftment and growth of each cell type for up to one month. In contrast, despite initial binding, all cell types inoculated into decellularized emphysematous lungs did not survive beyond one week. However, cell attachment and proliferation on solubilized ECM homogenates of decellularized normal and emphysematous lungs coated onto tissue culture plates was comparable and not impaired, suggesting that the 3-dimensional decellularized emphysematous scaffolds may lack the necessary ECM architecture to support sustained cell growth.     

Interaction of Pseudomonas aeruginosa with human lung fibroblasts: role of bacterial elastase.

Colonization of cell surfaces by Pseudomonas aeruginosa is mediated by bacterial adherence, which, in turn, is influenced by both host and microbial factors. Previous studies with this organism suggest that elastase contributes to tissue invasion and necrosis. We studied the effects of Pseudomonas elastase (PE) on the adherence of P. aeruginosa to human lung fibroblast monolayers. Treatment of fibroblasts with PE (1 microgram/ml or 0.06 U/ml) increased adherence of 35S-labeled P. aeruginosa to cells, but heat-inactivated PE did not affect bacterial adhesion. Immunocytochemistry of cultured cells showed that PE (0.06 to 0.63 U/ml) decreased fibronectin (Fn) on the cell surface and extracellular matrix of cultured human lung fibroblasts. Data obtained by cytofluorography indicated that elastase also decreased Fn receptors on fibroblasts. Additional evidence for Fn degradation was provided by SDS-PAGE analysis of soluble Fn and proteins from surface iodinated cell monolayers treated with PE. We conclude that the increased bacterial adherence to fibroblasts may be due, in part, to elastase-induced proteolysis of Fn and its receptors on cell surfaces. Degradation of Fn could thus influence the extent and course of Pseudomonas infection in the lungs.     

Human anti-fibronectin antibodies in systemic lupus erythematosus: occurrence and antigenic specificity.

Sera from patients with connective tissue diseases exhibit autoantibodies to a spectrum of extracellular matrix proteins. Antibodies binding to solid phase bovine fibronectin (Fn) were investigated by ELISA in patients with systemic lupus erythematosus (SLE). Sera showing binding of 2 s.d. above the mean of the normal human control were considered positive and 43/150 SLE sera (28.7%) demonstrated such binding. These antibodies were mainly IgG and IgA as determined by isotype-specific ELISA. Specificity studies on selected positive sera revealed that binding was inhibited by preincubation with soluble Fn, but not with thyroglobulin or type 1 collagen. The binding was demonstrated not to be related to interactions with rheumatoid factor, complement components or immune complexes. Additional studies to determine which Fn fragment is bound by naturally occurring anti-Fn antibodies demonstrated that the binding was predominantly to the 30-kD collagen binding domain (CBD) of Fn molecule. Inhibition studies using 120-kD, 40-kD and 30-kD Fn fragments confirmed that this binding site was the 30-kD CBD.     

Accumlation of Hyaluronate in Human Lung Carcinoma As Measured by A new Hyaluronate Elisa

We have developed a new ELISA to quantify hyaluronate. This ELISA takes advantage of an anti-keratan sulfate antibody to differentiate between the coated aggregating rat chondrosarcoma proteoglycan which captures the hyaluronate and the keratan sulfate-bearing aggregating proteoglycan added subsequently. Absorbance in this assay was shown to be linear to the logarithmic concentration of hyaluronate in the range of 15 to 1000 ng/ml. Pre-treatment of hyaluronate with papain or protease did not interfere with its quantification; in contrast, pre-treatment with 0.1N NaOH significantly interferes with the subsequent measurement of the hyaluronate molecules. The size of the hyaluronate molecule was found to be an important factor in quantification: only large size hyaluronate molecules could be quantified accurately. The ELISA was used to show that human lung carcinomas contain 2 to 500 times as much hyaluronate as normal lung tissue from the same patient.     

Role of pulmonary diseases and physical condition in the regulation of vasoactive hormones

Lungs have many non-respiratory metabolic functions, of which some take place in the capillary endothelium, while others are in parenchymal lung tissue. We have studied the role of the lungs in the metabolism of vasoactive and some other hormones by comparing patients who have undergone lung resection to those having various obstructive or fibrotic lung diseases. We have also compared these groups with persons in good physical health. The data suggested that lung resection patients had low angiotensin II levels in plasma but the response of angiotensin II to exercise was normal. Also adrenalin concentration was low in the lung resection group while dopamine did not show any significant difference between the groups. When hormone levels were correlated to the exercise data, renin levels were especially related to physical condition. Serum post-exercise renin values were inversely related to the uneven distribution of lung perfusion, possibly thus reflecting the diminished pulmonary vascularization. A negative association was found between angiotensin II and diffusion capacity. Thus, the angiotensin II levels may preferably be controlled by the non-circulatory functions of the lungs.     

Cellular FLICE-like Inhibitory Protein Deviates Myofibroblast Fas-Induced Apoptosis Toward Proliferation during Lung Fibrosis

A prominent feature of fibrotic tissue in general and of lungs in particular is fibroblast proliferation and accumulation. In patients overcoming fibrosis, apoptosis limits this excessive cell growth. We have previously shown resistance to Fas-induced apoptosis of primary lung fibroblasts from mice with bleomycin-induced lung fibrosis, their escape from immune surveillance, and continued accumulation in spite of overexpression of the Fas death receptor. Cellular FLICE-like inhibitory protein (c-FLIP) is a regulator of cell death receptor-induced apoptosis in many cell types. We aimed to determine c-FLIP levels in myofibroblasts from fibrotic lungs and to directly assess c-FLIP's role in apoptosis and proliferation of primary lung myofibroblasts. c-FLIP levels were determined by apoptosis gene array, flow cytometry, Western blot, and immunofluorescence before and after down-regulation with a specific small interfering RNA. Apoptosis was assessed by caspase cleavage in Western blot and by Annexin V affinity labeling after FACS and tissue immunofluorescence. Proliferation was assessed by BrdU uptake, also using FACS and immunofluorescence. We show that myofibroblasts from lungs of humans with idiopathic pulmonary fibrosis and from bleomycin-treated versus normal saline-treated mice up-regulate c-FLIP levels. Using the animal model, we show that fibrotic lung myofibroblasts divert Fas signaling from apoptosis to proliferation and that this requires signaling by TNF receptor-associated factor (TRAF) and NF-κB. c-FLIP down-regulation reverses the effect of Fas activation, causing increased apoptosis, decreased proliferation, and diminished recruitment of TRAF to the DISC complex. This indicates that c-FLIP is essential for myofibroblast accumulation and may serve as a potential target to manipulate tissue fibrosis.     

Enhanced mast cell chymase expression in human idiopathic interstitial pneumonia

Previous studies have shown that mast cell chymase induces and promotes fibrogenesis in injured tissues. We studied the roles of mast cell chymase in the fibritic processes of human idiopathic interstitial pneumonias. Frozen tissue sections from human lungs with usual interstitial pneumonia (n=7), nonspecific interstitial pneumonia (n=4) and normal lungs (n=10) were studied immunohistochemically. Monoclonal antibodies against mast cell chymase, tryptase, interleukin-4, and smooth muscle actin were used. Stained cells or areas were quantified by computer-aided morphometry. The numbers of both tryptase-positive mast cells and chymase-positive mast cells were significantly greater in lung tissues with idiopathic interstitial pneumonia than in normal lung tissues. The increase in the number of chymase-positive mast cells in the diseased lung tissues was closely related to an increase in interleukin-4-positive cells, and also to an accumulation of smooth muscle cells and myofibroblasts. Because smooth muscle cell and myofibroblast proliferation is a principal pathological change in idiopathic interstitial pneumonias, these observations suggest that mast cell chymase, possibly induced by interleukin-4-dependent phenotypic modulation, may be an important mediator in the inflammatory and fibrotic processes of idiopathic interstitial pneumonia in humans.     

Accumulation of hyaluronate in human lung carcinoma as measured by a new hyaluronate ELISA

We have developed a new ELISA to quantify hyaluronate. This ELISA takes advantage of an anti-keratan sulfate antibody to differentiate between the coated aggregating rat chondrosarcoma proteoglycan which captures the hyaluronate and the keratan sulfate-bearing aggregating proteoglycan added subsequently. Absorbance in this assay was shown to be linear to the logarithmic concentration of hyaluronate in the range of 15 to 1000 ng/ml. Pre-treatment of hyaluronate with papain or protease did not interfere with its quantification; in contrast, pre-treatment with 0.1N NaOH significantly interferes with the subsequent measurement of the hyaluronate molecules. The size of the hyaluronate molecule was found to be an important factor in quantification: only large size hyaluronate molecules could be quantified accurately. The ELISA was used to show that human lung carcinomas contain 2 to 500 times as much hyaluronate as normal lung tissue from the same patient.     

A total fibrinogen deficiency is compatible with the development of pulmonary fibrosis in mice

In addition to their well-known roles in hemostasis, fibrinogen (Fg) and fibrin (Fn) have been implicated in a number of other physiological and pathophysiological events. One of these involves the fibroproliferative response after acute lung injury, which is the focus of the current study. Mice with a total Fg deficiency (FG(-/-)) were generated by breeding heterozygous (FG(+/-)) pairs, each of which contained an allele with a targeted deletion of its Fg-gamma-chain gene. The resulting FG(-/-) animals did not possess detectable plasma Fg. FG(-/-) mice were then used to assess the roles of Fg and Fn in a bleomycin-induced acute lung injury model. Intratracheal administration of bleomycin in wild-type and FG(-/-) mice resulted in equivalent deposition of interstitial collagen and fibrotic lesions at days 7 and 14 after administration. This indicates that Fg and/or Fn are not essential for the development of bleomycin-induced pulmonary fibrosis.     

Disseminated intravascular coagulation (DIC) induced by liquoid (polyanetholsulfonate) in the rat. V. Effects on circulating fibronectin

Levels of plasma fibronectin (Fn) were 63% lower than normal 15 min after the intravenous injection of liquoid (P less than 0.01); 3 h later they were still low but rebounded to 35% above normal (P less than 0.01) by 24 h. Concurrently microthrombi containing fibrinogen, Fn and Factor VIII related-antigens (VIII:Ag) were detected in the kidneys and lungs by immunohistopathological studies. Ultrastructurally, thrombi were composed of dense granular and occasional fibrillar non-striated material. In liquoid-injected rats 125I-fibrinogen mainly localized in kidneys and lungs, especially in the latter (P less than 0.01), and the lungs had a higher wet-to-dry weight ratios than did controls (P less than 0.01). It is concluded that the polyanion (liquoid)-induced intravascular coagulation-like reaction sequestered Fn concomitantly with the precipitation of fibrinogen and VII:Ag in the microclots. The reduced concentration of plasma Fn may have impaired the disposal of coagulation products thus enhancing the expression of the coagulopathy-mediated renal and pulmonary histopathology. It is suggested that the liquoid-related coagulopathy may have resulted in enzymatic lysis of Fn.     

Disseminated intravascular coagulation (DIC) induced by liquoid (polyanetholsulfonate) in the rat. V. Effects on circulating fibronectin.

Levels of plasma fibronectin (Fn) were 63% lower than normal 15 min after the intravenous injection of liquoid (P less than 0.01); 3 h later they were still low but rebounded to 35% above normal (P less than 0.01) by 24 h. Concurrently microthrombi containing fibrinogen, Fn and Factor VIII related-antigens (VIII:Ag) were detected in the kidneys and lungs by immunohistopathological studies. Ultrastructurally, thrombi were composed of dense granular and occasional fibrillar non-striated material. In liquoid-injected rats 125I-fibrinogen mainly localized in kidneys and lungs, especially in the latter (P less than 0.01), and the lungs had a higher wet-to-dry weight ratios than did controls (P less than 0.01). It is concluded that the polyanion (liquoid)-induced intravascular coagulation-like reaction sequestered Fn concomitantly with the precipitation of fibrinogen and VII:Ag in the microclots. The reduced concentration of plasma Fn may have impaired the disposal of coagulation products thus enhancing the expression of the coagulopathy-mediated renal and pulmonary histopathology. It is suggested that the liquoid-related coagulopathy may have resulted in enzymatic lysis of Fn.     

ELISA方法测定纤维连结蛋白的临床应用

本文报告用抗人纤维连结蛋白单克隆抗体建立的ELISA方法对三组病人体内Fn水平测定的结果。实验结果表明肺部肿瘤、炎症时血浆和胸水内Fn较正常人明显为高,尤以肺癌为著,肝脏疾患和截瘫病人体内Fn水平低于正常人。反映了该两组病人肝功能和免疫功能的下降。     

E-cadherin基因在肺癌发生过程中的作用分析

目的检测E-cadherin在肺癌组织、癌旁组织及正常肺组织中的表达,并探讨其在肺癌发生发展中的作用。方法采用免疫组化SP法对22例肺癌组织、相应的癌旁组织和9例肺部良性病变旁正常肺组织中E-cadherin的表达状况进行检测。结果22例肺癌组织中E-cadherin蛋白表达减弱或缺失率59.1%(13/22)。明显高于相应癌旁组织中蛋白表达减弱或缺失率27.3%(6/22);9例正常肺组织中E-cadherin均正常表达。结论E-cadherin表达减弱或缺失是肺癌发生中的常见分子事件,可能参与了肺癌的发生发展过程。     

The influence of anti-fibronectin antibodies on interactions involving extracellular matrix components and cells: a possible pathogenic mechanism.

Antibodies, directed to the 30-kD collagen binding domain (CBD) of fibronectin (Fn), have been previously demonstrated in sera from patients with systemic lupus erythematosus (SLE), and we now investigate the possible pathogenic effects of these antibodies on collagen-Fn and cell-Fn interactions. The binding of type 1 collagen to Fn was demonstrated by ELISA, and could be specifically inhibited by the preincubation of solid-phase immobilized Fn with anti-Fn antibodies from SLE sera. By using indirect immunofluorescent staining, anti-Fn antibody containing SLE sera but not normal human serum (NHS) reduced the deposition of newly synthesized collagen and Fn on living human skin fibroblasts. We also found that sera from SLE patients containing anti-Fn antibodies significantly reduced thyroid cell attachment to Fn immobilized on plastic compared with NHS. These effects were shown to be due to the presence of anti-Fn antibodies in these sera, as SLE sera depleted of anti-Fn antibodies did not reduce the deposition of collagen or Fn on cultured fibroblasts, nor did they inhibit cell attachment.     

Anchoring fibrils. A new connective tissue structure in fibrotic lung disease

Electron microscopic studies of lung were made and compared in 17 patients with lung disease (10 with idiopathic pulmonary fibrosis, 3 with collagen--vascular diseases, 3 with sarcoidosis, and 1 with chronic eosinophilic pneumonia) and in 5 control patients. In control patients, the alveolar epithelial cells were normal, and no hemidesmosomes were present between the plasma membranes and the basal laminae. In comparison, cuboidal alveolar epithelial cells were present in 15 of the patients with fibrotic lung disease; in 9 of these the alveolar epithelial cells were multilayered. In 7 of the latter 9 patients (5 with idiopathic pulmonary fibrosis and 2 with collagen-vascular diseases), the basal laminae of the alveolar epithelial cells were attached to the plasma membranes by hemidesmosomes and to the underlying interstitial connective tissue by "anchoring fibrils." These fibrils measured from 4000 to 6000 A in length and from 200 to 600 A in width. One or both ends of the anchoring fibrils inserted into thebasal lamina, often forming arcs through which collagen fibrils and connective tissue microfibrils penetrated. Anchoring fibrils showed a complex pattern of transverse banding, which differed from that of collagen and appeared to be symmetric about the center of the fibril. These anchoring fibrils, which resemble those in normal skin and other tissues, were not found in lungs of control patients. In addition, there was a significant correlation between the severity of the pulmonary fibrosis and the presence of anchoring fibrils. These observations suggest that in severe fibrotic lung disease, anchoring fibrils reinforce the attachment of the basal lamina of multilayered alveolar epithelial cells to interstitial connective tissue.     

Fibroblast activation protein: a serine protease expressed at the remodeling interface in idiopathic pulmonary fibrosis

Fibroblast activation protein (FAPalpha) is a member of the cell surface dipeptidyl peptidase (DPP) family of serine proteases. In its dimer form, FAPalpha exhibits gelatinase, collagenase, and DPP activity in vitro. Reactive fibroblasts in healing wounds and stromal fibroblasts associated with epithelial tumors express FAPalpha. Idiopathic pulmonary fibrosis (IPF) is a disease of the lung characterized by progressive fibrosis with no clear etiology or molecular marker for disease activity. Recently, it has been shown that fibroblast FAPalpha expression is induced in liver cirrhosis, with an expression pattern distinct from alpha-smooth muscle actin (alpha-SMA). In this study, we determine whether FAPalpha expression is selectively induced in areas of ongoing tissue remodeling characterized by fibroblast foci in IPF. Human lung tissue was obtained from patients with IPF, centrilobular emphysema, and normal lung. Immunohistochemical studies were performed using anti-FAPalpha antibody and antibodies against alpha-SMA and CD26 (DPPIV), another member of the DPP family. We found that FAPalpha was not expressed in normal human lung tissue or tissue with evidence of centriacinar emphysema, but was induced in all patients with IPF and With a pattern distinct from that of CD26 found primarily on hyperplastic alveolar epithelium. Specifically, FAPalpha was detected in fibroblast foci and in fibrotic interstitium and not in the interstitium of adjacent architecturally normal lung. Alveolar/airway epithelium and vascular smooth muscle did not express FAPalpha. This is the first report of FAPalpha expression in IPF and our results suggest that FAPalpha is selectively induced in fibrotic foci, but not in normal or emphysematous lung. Future studies will address whether FAPalpha may be used as a marker for disease activity in IPF.     

Decreased asbestos-induced lung inflammation and fibrosis after radiation and bone marrow transplant

The effect of lung irradiation on subsequent inflammatory or fibrotic lung injuries remains poorly understood. We postulated that irradiation and bone marrow transplantation might impact the development and progression of lung remodeling resulting from asbestos inhalation. Our objective was to determine whether irradiation and bone marrow transplantation affected inflammation and fibrosis associated with inhaled asbestos exposure. Inflammation, cytokine production, and fibrosis were assessed in lungs of na?ve and sex-mismatched chimeric mice exposed to asbestos for 3, 9, or 40 days. Potential engraftment of donor-derived cells in recipient lungs was examined by fluorescence in situ hybridization and immunohistochemistry. Compared with asbestos-exposed na?ve (nonchimeric) mice, chimeric mice exposed to asbestos for 3, 9, or 40 days demonstrated significant abrogation of acute increases in asbestos-associated inflammatory mediators and fibrosis. Donor-derived cells trafficked to lung but did not significantly engraft as phenotypic lung cells. Irradiation and bone marrow transplantation alters inflammatory and fibrotic responses to asbestos, likely through modulation of soluble inflammatory mediators.     

Presence of platelet-derived growth factor in normal and fibrotic lung is specifically associated with interstitial macrophages, while both interstitial macrophages and alveolar epithelial cells express the c-sis proto-oncogene

Normal lung structure is maintained by the presence of mesenchymal cells and their extracellular matrix products. The slow normal turnover of these cells is disrupted in fibrotic disorders, resulting in the in situ accumulation of mesenchymal cells and their extracellular matrix leading to a progressive alveolar wall thickening. Idiopathic pulmonary fibrosis (IPF) is a chronic fibrotic disorder of the lung characterized by a diffuse interstitial and intra-alveolar inflammation dominated by macrophages and polymorphonuclear neutrophils. Evaluation of alveolar macrophages (AM) obtained by bronchoalveolar lavage has previously shown that AM from normal individuals spontaneously release small amounts of platelet-derived growth factor (PDGF), a chemotactic and growth factor for mesenchymal cells, whereas AM from IPF patients spontaneously release increased amounts of biologically active PDGF, suggesting its involvement in mesenchymal cell accumulation. However, other cells such as endothelial cells and vascular smooth muscle cells can also release PDGF in vitro. In order to specify PDGF location in lung parenchyma, open lung biopsies from normal individuals and IPF patients were examined by immunohistochemistry using an anti-PDGF antibody and by in situ hybridization using PDGF A-chain and B-chain gene probes. In normal as well as in fibrotic lung, PDGF was only present in relation with interstitial macrophages but not with any other inflammatory cells or mesenchymal cells. Furthermore, the percentage of PDGF-positive macrophages in IPF was 3-fold increased in comparison to normal lung. In addition, the percentage of PDGF-positive macrophages was the same in fibrotic and nonfibrotic areas of IPF lungs.(ABSTRACT TRUNCATED AT 250 WORDS)     

Perfusion with lipopolysaccharide negative blood eliminates lipopolysaccharide induced lung injury

We investigated whether perfusion with control blood improves pulmonary functions compromised by lipopolysaccharide (LPS) infusion. This was an animal study in a research laboratory at a university hospital by using Sprague-Dawley rats (n = 19), each weighing 325 to 350 g. All animals were pretreated with a 24 hour infusion of either LPS (5 mg/kg) or vehicle, after which, excised lungs were reperfused for 2 hours with either LPS+ or control blood. Three groups were studied: (1) group S (n = 6); LPS pretreated lungs reperfused with LPS containing blood to mimic persistent sepsis, (2) group N (n = 6); LPS pretreated lungs reperfused with control blood to mimic the removal of the septic blood components, and (3) group C (n = 7); vehicle pretreated lungs reperfused with normal blood as a control. Blood gas exchange, shunt fraction (Qs/Qt), alveolar-arterial oxygen gradient (A-aDO2), and variables for lung mechanics were measured. Leukosequestration was quantified with a myeloperoxidase (MPO) assay. The PO2 (mm Hg) values at 90 min after reperfusion in groups S, N, and C were 67.8 +/- 7.0*, 85.2 +/- 9.2, and 90.1 +/- 7.5, respectively (*p < 0.05; vs. group N and C). In addition to PO2, A-aDO2 and Qs/Qt significantly deteriorated in group S. MPO activity in the lungs after LPS infusion was significantly higher than that after vehicle infusion (1.7 +/- 0.3 vs. 0.12 +/- 0.04 units/g tissue; p < 0.001). Subsequent reperfusion with LPS+ blood (group S) increased MPO activity to 3.1 +/- 0.6 (p < 0.05), but reperfusion with normal blood (group N) caused a significant decrease to 1.1 +/- 0.2 (p < 0.05). MPO activity in group C did not significantly change compared with those after vehicle infusion. Reperfusion with control blood normalized lung function compromised by pretreatment with LPS and significantly reduced leukosequestration. These results favor the possibility that the removal of LPS+ blood components may eliminate septic lung injury.